Despite the widespread abuse of ethanol, the neurochemical mechanisms by which ethanol induces self administration and dependence are still poorly understood. Although multiple pharmacological actions are probably involved in the effects of ethanol, there is strong evidence that some of its pharmacological properties are mediated by its interaction with the GABA-benzodiazepine receptor complex (GBRC). For example, both ethanol and benzodiazepines (bdz) enhance GBRC mediated inhibitory neurotransmission by facilitating Cl- flux at the GBRC. Decreased inhibitory neurotransmission as a consequence of functional changes in the GBRC in response to chronic ethanol could contribute to alcohol dependence and/or withdrawal. Alternatively genetic differences in the expression of the GBRC could influence an individuals response to ethanol and/or bdz and could contribute to their abuse of these drugs. Based on recent PET studies in our laboratory documenting in alcoholics a decreased response to changes in regional brain glucose metabolism induced by bdz we hypothesize that alcoholics have an abnormal response to GBRC mediated inhibitory neurotransmission. This proposal will test this hypothesis and will evaluate the extent to which the decreased response to GBRC mediated inhibitory neurotransmission represents a temporary rather than a long lasting change. GBRC mediated inhibitory neurotransmission will be indirectly examined by measuring the regional brain metabolic response to a challenge with a bdz drug using PET and 2-deoxy-2[18F]-fluoro-D-glucose. Serial PET studies will be done before and after challenge with lorazepam (30 mu g/kg) in 30 alcoholics and 20 normals. The short-term effects of alcohol withdrawal will be distinguished from long-term effects by testing the alcoholics at 2 and at 8-10 weeks after alcohol withdrawal. The normal controls will be tested over the same time period to control for intrasubject variability. Our working hypotheses are: 1) alcoholics have a decreased response to GBRC mediated inhibitory neurotransmission (which will be reflected in a blunted response to the regional brain metabolic effects of lorazepam); 2) the decreased response to GBRC mediated inhibitory neurotransmission is not a temporary response to ethanol discontinuation (the blunted response to lorazepam will persist after withdrawal); 3) decrements in whole brain metabolism during early detoxification in alcoholics will recover with protracted withdrawal. Preliminary data support these hypotheses. Abnormalities in GBRC mediated inhibitory neurotransmission in alcoholics could be one of the mechanisms underlying ethanol dependence.